The chronic depressive disorders, which afflict 3 percent to 5 percent of the adult population of the U.S., are potentially disabling conditions that adversely affect physical health and performance in social and vocational roles. Research conducted by our team since 1991 has documented 50 percent acute phase intent-to-treat response rates to standard antidepressants and 75 percent prophylaxis with maintenance therapy. Many experts have suggested that a combination of psychotherapy and pharmacotherapy may be the optimal treatment for chronic depression and, indeed, we have completed a study (n=681) in which the combination of nefazodone and a model of psychotherapy developed for chronic depression (Cognitive Behavioral Analysis System of Psychotherapy; CBASP) was significantly more effective than either monotherapy (intent-to-treat response rates: 75 percent vs. 48 percent vs. 45 percent). These findings have limited generalizability because of the highly restrictive nature of the protocol. Moreover, it is not clear if the more expensive combined approach ultimately results in a greater proportion of fully recovered patients, especially when compared to an optimized sequential, algorithm-guided approach to pharmacotherapy. It also appears that while CBASP alone may have a slower onset of action than pharmacotherapy alone, those who do respond to this form of psychotherapy have a more durable or enduring response. We now propose to extend these findings in an innovative multicenter trial of a diverse and representative sample of 450 chronically depressed patients, The study will encompass both acute and continuation phases of treatment. During the acute phase (lasting up to 28 weeks), patients will be randomized to one of three modalities: 1) sequential algorithm-guided pharmacotherapy (n=150), 2) CBASP alone (n=150), or 3) pharmacotherapy plus CBASP (n=150). During both the acute and continuation phase trials we predict that combined treatment will yield superior outcomes on symptom, syndromal (i.e., remission, recovery, and relapse rates), and functional measures. Across the full study duration, the treatments will be contrasted on a variety of quality of life and health economic outcomes. We predict that the extra cost of combined treatment will be at least partly offset by faster and more complete remissions. We also predict pharmacotherapy alone will have faster effects (at lower costs) than CBASP alone during the acute phase but that the CBASP group will "catch up" by the end of the continuation phase. This two stage study bridges conventional efficacy and services models of research and will make a significant impact on public health by evaluating both the utility and cost of these well-characterized treatment approaches. The study also will provide the foundation for a subsequent application evaluating longer term, maintenance phase treatments of chronic depression.